=== salem_ is now known as _salem [01:05] alecu: ping [01:50] alecu: bug 1466273 [01:50] bug 1466273 in Thumbnailer "gstreamer fails intermittently" [Undecided,New] https://launchpad.net/bugs/1466273 [02:05] michi: thanks! hope everything goes alright with your achilles heel [02:05] alecu: I shall live. Sorry, I posted this in the wrong channel by mistake. [02:05] Meant to send it in unity-api. === ljp is now known as lpotter [04:54] morning [07:44] jibel, davmor2: since I had my evening off yesterday, I probably missed the ping... but did the ubuntu images pass sanity? [08:11] sil2100, it did [08:12] jibel: excellent, so I can promote those, right? [08:12] jibel: QA tested mako, or any other platform too? [08:14] sil2100: yes you missed the ping [08:14] sil2100, just mako [08:14] sil2100: you also missed the cussing for there being a 164 when I went to test it too :P [08:15] davmor2: oh, uh, then good I missed that I guess! I made 164 since people needed shell rotation back [08:15] hm, I wonder how broken flo and manta are [08:17] sil2100: yeah only I didn't know about it so install expecting 163 fortunately I notice the new icons and shell rotation so look into what was wrong [08:18] sil2100: is manta even supported anymore? [08:18] I mentioned the image number which was the OTA-4, so I thought you would install it explicitly :) [08:18] Well, no one told us not to support it, we still build images for it [08:19] The discussion never got resumed about deprecating it [08:19] Maybe it's a good time to start that agin [08:19] *again [08:22] For now I'll promote mako, generic* and krillin [08:34] Yeah... we anyway don't seem to have OTA-4 images for flo, manta and generic_x86 right now... [08:34] Might need to think of a smart way to make them [08:37] sil2100: i added a line to the sheet for clock but I see no trello card, am I doing something wrong? [08:38] popey, are the system connected now? [08:38] like adding a line to the gdoc talks to trello? [08:38] i was told there was a bot [08:38] interesting ;-) [08:38] seb128, since forever (for the "testing done" filed at least) [08:39] ogra_, only when QA is required no? [08:39] the click package stuff not as long though [08:39] seb128, yeah [08:39] k, right [08:39] popey, the landing team has to assign a landing id. [08:39] I just did, it should appear on the board soon [08:40] popey, well, when it'll be ready for QA [08:40] ok. === chihchun_afk is now known as chihchun [08:41] popey, if you tested it, mark it so and it should be enough to set it ready for qa [08:44] sil2100: FYI QA now has a Qt security update in their hands... not sure if security updates to overlay contents have been discussed or not, but this relied on me noticing it (vivid proper was updated but it does not affect our images) [08:46] sil2100, https://wiki.ubuntu.com/LandingTeam/SupportedDevices doesn't mention flo or manta for stable channels [08:46] jibel: ah, you're right [08:47] jibel: we decided that we'll only do them for the devel channels [08:47] My bad, forgot about that [08:48] Mirv, security updates should follow the same process than any other security updates independently of the target archive/ppa [08:48] jibel: what I meant as a general topic is that our security team does not currently consider overlay PPA at all [08:49] so if we have a component there, it does not get security attention unless the component owner does so [08:49] Mirv: who is working on the Qt security update? [08:50] sil2100: ? [08:50] sil2100: so as I just said I did this overlay update now ready for signoff [08:50] sil2100: security team did it for normal vivid two weeks ago [08:50] and I had a newer version in wily already [08:51] Mirv: ah, ok, and the reason why we didn't get the security update 2 weeks ago is because we have our own Qt in the overlay [08:51] Ok [08:51] sil2100: so I just raised the topic that currently our security related things are not consolidated, as security team handles only stable Ubuntu releases but not phone if a package happens to be in overlay ppa [08:52] sil2100: yeah, that :) [08:52] Mirv: yeah, we know about it, and the reason it's like that because by principle anything that lands in the overlay should be backported as an SRU to vivid (if possible) [08:52] then if we use SRU style versioning we run into the same thing as network-manager that overlay PPA package might be overwritten by normal SRU and losing overlay commits [08:53] Well, that was the idea in the past, I know slangasek and olli are still re-discussing that [08:53] sil2100: riiight, yes, I remember this now. [08:53] sil2100: ok, let's see later. so now we have examples of Qt that needs manual intervention and network-manager that might suddenly lose features on phone images - two approaches [08:53] For Qt the situation is anyway a bit harder, as it's not a touch-specific component [08:54] sil2100: so is not network-manager, so that's another good example [08:54] Yeah... [08:54] and some of the indicators too [08:55] and UITK [08:55] and webbrowser [08:55] these latter examples are such that have actual bumped upstream versions so they are essentially same as Qt [09:42] sil2100, did anything change recently with the spreadsheet, the bot cannot parse the click tab anymore [09:42] ? [09:46] jibel: hm, nothing from our side at least [09:46] jibel: we did notice that the spreadsheets have changed, so maybe they modified the API somehow? [09:48] sil2100, I'm just doing a get passing the id of the tab in the url but now it always returns the 1st tab [09:50] You doing it as CSV or using the API for that? === chihchun is now known as chihchun_afk [09:51] sil2100, CSV, the format of the url has changed [09:51] fixing no [09:51] w [09:52] Ok, probably due to the switch to the new API [09:53] Actually, I don't want to jinx it, but maybe they fixed the race conditions in the new version and we won't have things broken anymore [09:57] fixed [09:57] popey, https://trello.com/c/kTTrqblZ/1802-click-clock-app-popey [09:58] thanks jibel [10:07] Mirv, silo 10 is the security fix you mentioned earlier? [10:15] \O/ [10:15] (big head since cat enabled capslock) [10:15] jibel: yes [11:09] wgrant, cjwatson: hey guys! I wanted to poke about the overlay translation updates - are all the bits in LP ready for that? Or is that still just a proposition and you're waiting for us? [11:09] It's a proposition. [11:10] If you folks sign off on it, we can implement it, but we weren't going to do it entirely on our own initiative. [11:18] cjwatson: ok, need to move this forward then, but I suppose it might be linked with the big derived-vs-overlay discussion [11:20] sil2100: That's what I'd feared. I think it would be a bad outcome to get stalled in some kind of perfect-is-the-enemy-of-the-good thing [11:21] sil2100: Just don't quite have the bandwidth to be able to give a day to something that might not be used [11:21] At best I would like to have this implemented without waiting on the decision, but if by any chance we decide that its best to switch to a derived distro (uh) then it owuld be wasted work [11:21] Yeah [11:22] cjwatson: ok then, we'll be in touch regarding that [11:23] We can do it if there's a decent chance === _salem is now known as salem_ === MacSlow is now known as MacSlow|lunch [12:02] * sil2100 off to lunch === alan_g is now known as alan_g|lunch [12:24] if cmake-extras is installed, why would EnableCoverageReport not be found?! === MacSlow|lunch is now known as MacSlow [12:51] trainguards: I don't like to nag, but if someone could push the land button on silo 43 it would probably help out quite a few people :) (anyone using cmake-extras) [12:52] I may or may not be one of these folks ;) [12:58] pete-woods: ok :) [12:59] pete-woods: unapproved! https://code.launchpad.net/~pete-woods/cmake-extras/cmake-3.2/+merge/262315 .. [12:59] (just top-approve if you have the authority) [12:59] Mirv: done! [13:00] pete-woods: ok, it's publishing now and I acked the packaging changes [13:01] Mirv: awesome, thanks! === mvo_ is now known as mvo === alan_g|lunch is now known as alan_g [13:53] robru: after cjwatson diagnosed the problem, i wonder if there isn't a quick hack/fix to add an option to include -sa when building the package. adding .bzr-builddeb doesn't work [13:53] (i can see that the orig.tar.gz isn't included in the upload to the ppa) [13:54] robru: or hmm... maybe i can just upload to the ppa manually and then let the train take it all from there? === mvo_ is now known as mvo [14:05] jibel: ping https://code.launchpad.net/~bzoltan/ubuntu-ui-toolkit/landing_12-6-2015/+merge/261831 the branch is top-approved now [14:05] (and tested) [14:05] (for silo 011) [14:12] o/ silo 001 ready to publish for wily; though some qa smoke testing might be good for the dual/vivid task as well [14:13] o/ [14:17] kalikiana, thanks [14:17] * jibel adds a task to update the bot to update existing cards. [15:12] barry: yeah you can always upload manually. I have no idea why your package alone requires -sa, literally no other train packages require this and it's never been a problem before. [15:14] robru: just trying to keep things interesting for you [15:16] robru: i'll upload manually for now [15:18] barry: would -sa option be a nop for packages that are already including orig.tar in the upload anyway? I'd be open to adding that to all builds if it a) helped you and b) was harmless for everybody else. I don't like the idea of adding extra options to the build, too many checkboxes already [15:19] robru: i think it would be harmless. i'm not actually sure how it's currently automagically including the orig.tar.gz or not, or why my branch isn't getting it [15:20] barry: my only guess is that it's because you have a debian/watch. nobody else does. without it, dput must think "there's no way to find the orig.tar so I have to include it" but with yours dput is like "oh yeah lp can find the orig.tar on it's own" [15:22] robru: hmm... but that would be an easy theory to test. what's the command used to create the source package from the mp? [15:24] barry: "bzr bd -S" [15:26] barry: makes me wonder why lp doesn't just use your debian/watch to find the orig.tar [15:27] robru: that is very strange, because it's the same command i use here, and my _sources.changes *does* have an orig.tar.gz. otoh, i'm doing it from the branch that already has the merge. let me try to branch the packaging branch and then merge in my mp, and do that command [15:46] 7~~~~~~~~~~~4~ [16:00] sil2100, meow ? [16:00] trainguards: can I have a silo for line 68, please? [16:01] landing meeting? [16:14] robru, slangasek yeah, so my source package building test worked just fine here locally. in a completely different, fresh directory i did the following: [16:14] bzr branch lp:~ubuntu-managed-branches/ubuntu-system-image/system-image [16:14] cd system-image/ [16:14] bzr merge lp:~barry/ubuntu-system-image/citrain301 [16:14] bzr bd -S [16:15] then... [16:15] % grep orig ../system-image_3.0.1-0ubuntu1_source.changes [16:15] 4fa47281da9e6f313c2ab27c83e129bf4ae747fe 203601 system-image_3.0.1.orig.tar.gz [16:15] 3a0d73bd37a67bcc1606d4f082d2072af5555f8413c36e56ad4b50f3b9ddb277 203601 system-image_3.0.1.orig.tar.gz [16:15] 447b49e1a59655a16451bc98543190f8 203601 python optional system-image_3.0.1.orig.tar.gz [16:15] [16:15] so yeah, i have no clue why the train is unhappy [16:15] for now, i am just going to upload manually to the ppa [16:16] robru: in the meantime, please remove the line that says "if user == 'barry': raise MysteriousFailureError" [16:17] barry: no i like that one! [16:17] robru: somehow that got past my code review [16:17] barry: I'll push a branch that adds -sa and see what happens [16:17] robru: should i wait so we can test that out for reals? [16:18] barry: sure one sec [16:20] barry: ok started a build [16:20] cool [16:20] robru: debian/watch is totally irrelevant for this [16:21] cjwatson: ok well I added -sa so we'll see if that works [16:21] -sa is implied if the previous version in the changelog had a different upstream, iirc [16:21] cjwatson: but it's bizarre to me that all other packages work without -sa, I'm not sure how barry's differs [16:21] robru: oops [16:21] robru: you need to put the -sa after the -- [16:21] barry: cjwatson but 3.0.1 differs from 3.0? [16:22] it does [16:22] cjwatson: yep, and locally no -sa is needed [16:22] hang on [16:23] robru: this doesn't look right: [16:23] Creating source with bzr bd -S -sa -- -k55EFCAA8 -d -v3.0-0ubuntu2 [16:23] first, the -sa needs to go after -- [16:23] but -v3.0-0ubuntu2? [16:23] that's probably breaking debuild [16:23] where's that coming from? [16:23] barry: no the -v tag means "include the diff since this version", that's the version currently in archive [16:23] barry: fixed the -sa though [16:24] robru: oh right of course [16:24] bingo [16:24] Uploading system-image_3.0.1.orig.tar.gz: done. [16:24] [16:25] nfc why that's not autodetected, but it looks like that might have solved the problem [16:25] barry: cool. I'll keep an eye on that to make sure nobody else gets broken by this change ;-) [16:25] :) [16:25] robru, can you please reconfigure silo 19 [16:25] using -sa all the time is inefficient but shouldn't actually be harmful [16:26] robru: it could be a problem if someone uploads a change that only affects the packaging version, but given that most are not split like s-i i bet that won't happen in practice like it does in the distro [16:26] barry: yeah in the train we auto-generate new upstream versions [16:26] barry: that just means the orig is reuploaded unnecessarily, but LP won't care as long as it's identical [16:27] cjwatson: ah, cool. thanks guys! i'm off to get some lunch but will check back in a few [16:28] it's in https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-040/+packages and building [16:28] but you can see there why there was a problem! [16:28] robru: look at https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-040/+packages and expand the expander [16:28] observe the duplicate changelog version [16:29] cjwatson: oh... k... [16:29] robru, thanks [16:29] cjwatson: I blame barry [16:29] jhodapp: you're welcome [16:29] that confuses dpkg-buildpackage into thinking it doesn't need to reupload the .orig because look there was a "previous" version already [16:30] (dpkg-genchanges, strictly) [16:30] cjwatson: other than looking retarded, is that going to cause any further problems? [16:30] shouldn't [16:30] alright [16:30] davmor2, how do i need to bug to get silo 4 verified by the eod of the week? [16:31] s/eod/end/ [16:31] kenvandine: not me I'm off tomorrow :) ↑ jibel [16:32] kenvandine: it is however only 3rd in the queue so will most likely be done tomorrow anyway [16:32] davmor2, i was told at one point that they aren't verified in that order [16:33] it was first in the queue a few days ago :) [16:33] kenvandine: yeah if something critical lands then everything else gets bumped one. but on the whole that is the order we pull from [16:34] ok [16:40] ricmm, you asked in the past to get a heads up on MPs for papi. This has lingered for a while now : https://code.launchpad.net/~mir-team/platform-api/delete-deprecations/+merge/254170 and it's blocking slangasek. Can you please take a look? We'll be landing it soon. [16:41] racarr, AlbertA ^ === alan_g is now known as alan_g|EOD [17:20] cjwatson: well, wtf. yay. course, it was happening even before that [17:20] well, wait a sec [17:20] my branch doesn't have that [17:21] and neither apparently does: https://code.launchpad.net/~barry/ubuntu-system-image/citrain301/+merge/262255 [17:21] cjwatson, robru ^^ [17:22] * barry blames robru's == 'barry' line [17:22] barry: yeah the train adds that because reasons. ideally you'd allow the train to generate versions for you and it would all be one big entry but because you force your own versions this happens [17:22] sweet [17:23] * barry thinks of more ways to break robru's software === mhall119_ is now known as mhall119 === alex-abreu is now known as alex-abreu|afk [18:32] cyphermox: happy birthday! ;-) [18:39] sil2100: why can't i find s-i 2.5.1 in the dashboard? [18:41] or robru ^^? [18:42] barry: 2.5.1? hm? [18:42] robru: row 55 [18:43] barry: because it was published: https://ci-train.ubuntu.com/job/ubuntu-landing-028-2-publish/42/console [18:44] barry: and it's in the PPA: https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/stable-phone-overlay/+packages?field.name_filter=system-image&field.status_filter=published&field.series_filter= [18:44] robru: then why row 55, column P? [18:44] is that just leftover artifact or is there Something To Do? [18:45] barry: seems stale. spreadsheet rubbish [18:45] barry: here's something to do: https://code.launchpad.net/~barry/ubuntu-system-image/citrain301/+merge/262255 please top approve [18:45] robru: cool. have you killed the spreadsheet yet? [18:45] it's close, barry. so close. [18:45] :) [18:45] done [18:46] barry: if you wanna help with spreadsheet, visit http://requests.ci-train.staging.ubuntu.com/ sign in, create a request, fiddle around, and file bugs against https://bugs.launchpad.net/bileto [18:48] robru: will do. can you actually build and publish packages there currently? [18:49] barry: not publish, no [18:49] barry: it's a staging instance, it can build packages in ~ci-train-staging-area ppas, which are virtualized, and thus not eligible for publishing. [18:49] robru: ack [18:53] * barry thinks that's exactly where system-image belongs [18:54] robru: one nice thing about bileto - i won't have to have a dual monitor spanning chrome tab open to interact with it [18:57] barry: yes one of the primary goals is to be more streamlined than a spreadsheet. [18:57] barry: it's not perfect, but it's an order of magnitude better [18:57] brb [19:20] trainguards, can I get a silo for row 69 [19:21] queuebot, indeed [19:21] robru: virtualised> soooo close [19:22] virt ppc64el builds almost work and we have hardware installed for ARM ... [19:23] charles: need MPs, not branches [19:23] cjwatson: sweeeeet [19:23] d'oh [19:23] cjwatson: still though even with that I'm not sure publishing from ci-train-staging-area is a good idea. the creds are less credible. [19:24] robru, corrected [19:24] charles: thanks [19:24] charles: silo 22 [19:25] robru: no quite [19:25] will make ephemeral PPAs simple though [19:25] (ish) [19:25] cjwatson: right, that'll be nice [19:25] robru, thanks [19:25] charles: you're welcome === alex-abreu|afk is now known as alex-abreu === salem_ is now known as _salem